JPS62124456A - Automatic calibration system for mass spectrometric analysis data - Google Patents

Abstract

PURPOSE:To form a corresponding table of an appearance time of peaks and known mass number based on the measurement data of a standard sample by providing a peak detecting means which detects the peak of theoretical mass number coinciding within the permissible error from the allocated peaks. CONSTITUTION:The measurement data 1 consisting of the time value and peak intensity obtd. by measuring the standard sample is subjected to the detection of the prescribed number of peaks in order of higher intensity by a peak extracting part 3. The theoretical mass number is calculated from the measurement conditions and the standard sample by a mass number calculating part 4. On the other hand, the known meass number of the standard sample is registered in a standard file 2. An allocating part 5 retrieves and allocates the peaks of the theoretical mass number within the permissible error with respect to the known mass number registered in the file 2. The three known mass numbers of the smaller samples are taken by a corresponding part 6 and the peaks coinciding with the permissible error are corresponding thereto from the peaks of the theoretical mass number allocated to the respective known mass numbers by which the corresponding table of the appearance time of the peaks and the known mass number is formed.

Description

[Detailed Description of the Invention] C. Industrial Application Field] The present invention automatically creates a correspondence table between the appearance time of a standard sample peak and known mass number, which is used to calculate the peak w amount number of mass spectrometry data. This invention relates to an autocalibration system for mass spectrometry data.

[Conventional technology]

FIG. 4 is a diagram showing an example of the configuration of a mass spectrometry data system,
Figure 5 is a diagram for explaining the data stored on the disk, Figure 6 is a diagram for explaining the contents of the standard file,
FIG. 7 is a diagram for explaining measurement conditions, FIG. 8 is a diagram showing an example of the structure of measurement data, and FIG. 9 is a diagram showing an example of a peak pattern.

In the figure, 11 is a mass spectrometer, 12 is an interface,
13 is a CPU, 14 is a memory, 15 is a CRT terminal, 16 is a graphic printer, and 17 is a disk.

As shown in FIG. 4, the mass spectrometry data system consists of a mass spectrometer 11, a CPU 13, a memory 14, a CRT terminal 15, a graph ink printer 16, a disk 17, etc., and the mass spectrometry data obtained by the mass spectrometer 11 is used as an interface. Memory 1 through 12 and CPU 13
Incorporated into 4. The disk 17 stores standard files, measurement conditions, measurement data, etc. as shown in the fifth figure. The standard file consists of the time value and peak intensity corresponding to known substance 1, as shown in Figure 6, and the measurement condition data includes the mass number at the start of scanning, the mass number at the end of scanning, and the mass number at the end of scanning, as shown in Figure 7. It consists of the time required for scanning, sampling time, scanning type, etc.

In addition, the measurement data includes the number of peaks, as shown in Figure 8.
Consists of time value and peak intensity.

As described above, basic data (measured data) obtained from a mass spectrometry data system includes the time and intensity at which a mass spectrum (hereinafter referred to as a peak) appears. A sample consisting of (
When a certain compound) is introduced into the mass spectrometer 11 and measured by the data system, the peak pattern unique to the sample is shown in Figure 9, where the horizontal axis is time and the vertical axis is intensity.
CRT terminal 15 through the system in the form of dimensional information
will be displayed. The time for each peak then corresponds to an unknown mass number. Therefore, in order to know the mass number of a peak, it is necessary to have a correspondence table between time and known mass number, from which the mass number of the peak can be calculated from the relational expression between time and mass number. A standard file stores this correspondence table, and the process of creating this correspondence table is called calibration.

In calibration, a sample with a known mass number (such as PFK and FOMBLINE, hereinafter referred to as standard samples) whose peak appearance interval 1i is constant such as 12 squares or 14 squares is measured and obtained. A person uses a terminal key or the like to specify the mass numbers of prominent peaks (at least three) with high intensity in the peak pattern through the system. After designation, the CPU assigns known mass numbers to the remaining peaks based on the relational expression between time and mass number, and the creation of the correspondence table is completed. Note that the peak intensity is also recorded in the standard file.

The above method only collects known mass numbers in the standard file.
, ★, and the corresponding time values and peak intensities are all 0, that is, unknown. Incidentally, as part of the calibration, there is a function to record only known mass numbers into a standard file G.

- There is also a method of performing gear represion on a newly measured standard sample using an existing standard file with force correspondence. In this case, the CPU selects 3 mass numbers for peaks with high intensity from the standard file, selects 40 peaks with high intensity on the measurement side, and determines which of the 40 peaks the 3 mass numbers correspond to, ′ The CPU searches to see if they correspond using a relational expression and an appropriate error range. Then, if there are three peaks that are associated with each other, a new standard file can be created.

[Problem that the invention seeks to solve]

By the way, in calibration using a standard file whose time and intensity are all unknown (hereinafter referred to as new calibration), as mentioned earlier, a human must always perform at least 3
I had to associate the known mass number with the measured peak of the tree. However, in this case, there is a problem that it is very difficult to specify the 3-tree when there are many peaks whose intensity overflows.

In addition, in calibration using the associated existing standard file (hereinafter referred to as update calibration), if any one of the three known mass numbers selected by the CPU does not exist in the measurement side peak 40 tree. If this happens, there is a problem that a standard file with incorrect mapping will be created.

The present invention solves the above-mentioned problems by extracting and determining peaks according to a predetermined algorithm and automatically matching peak appearance times and known mass numbers based on measurement data of four samples. The purpose is to provide an autocalibration system for mass spectrometry data that can create tables.

[Means for solving problems]

To this end, the mass spectrometry data autocalibration system of the present invention includes a peak extraction means that extracts a predetermined number of peaks in order of intensity from measurement data of a standard sample, and a mass number calculation that calculates the theoretical mass number of the extracted peaks. means,
A standard file that registers the known mass numbers of standard samples, a peak assignment means that assigns peaks with theoretical mass numbers that are within the tolerance of the known mass numbers, and a peak assignment method that assigns the peaks of the three known mass numbers in order from the smallest mass number. The present invention is characterized in that it includes a peak search means for searching for a peak with a theoretical mass number that matches within an allowable error from the peaks obtained, and the appearance time of the searched peak is made to correspond to the known mass number in the correspondence table.

[Effect]

Auto-calibration of mass spectrometry data of the present invention
The system measures 4 standard samples, provides the measurement data and measurement conditions, and registers the known mass number of the standard sample in the standard file. Automatically extracts peaks from measurement data, calculates the theoretical mass number of each peak, assigns peaks with known mass numbers and theoretical mass numbers, and searches for peaks corresponding to known mass numbers without specifying peaks. A correspondence table is created.

〔Example〕

Examples will be described below with reference to the drawings.

FIG. 1 is a diagram showing the configuration of one embodiment of the autocalibration system for mass spectrometry data according to the present invention, FIG. 2 is a diagram for explaining the flow of processing by the autocalibration system, and FIG. FIG. 3 is a diagram showing an example of display output.

In Figure 1, 1 is measurement data, 2 is a standard file,
3 is a peak detection section, 4 is a mass number calculation section, 5 is an allocation section, 6
indicates the correspondence part. Measurement data 1 is data consisting of time values and peak intensities obtained by measuring a standard sample as shown in Figure 8, and standard file 2 is data consisting of known mass numbers of the standard sample as shown in Figure 6. is registered. The peak extraction unit 3 extracts a predetermined number of peaks from the measurement data 1 in descending order of intensity, and includes a WNN counter phloem 4,
For each peak extracted by the peak extractor 3, the seventh
The theoretical mass number is calculated from the measurement conditions and time values shown in the figure. The assignment unit 5 searches for and assigns peaks with theoretical mass numbers that are within the allowable error for the known mass numbers registered in the standard file 2, and the correspondence unit 6 searches for and assigns peaks with theoretical mass numbers that are within tolerance to the known mass numbers registered in the standard file 2. Take the three known mass numbers from
This is to associate peaks that match within an allowable error from the peaks of the theoretical mass numbers assigned to each known mass number.

Conventionally, as mentioned above, new calibration and updated calibration required manual designation and peak intensity reference of existing standard files, but the auto-calibration of the present invention eliminates this need. In other words, the autocalibration system for mass spectrometry data according to the present invention automatically calibrates the measurement data of the standard sample if a standard file containing only known mass numbers is created using the configuration shown in Figure 1. This is to correlate the known mass number with time.

Below, we will explain the specific process flow of this system in the second section.
This will be explained with reference to the figures.

"Step (a)" Measure the standard sample and file the measured data on a disk. At this time, in addition to creating a file including the measurement conditions, the θ-resistance data with almost the same beak deflection was obtained by continuous scanning. ! Make several and number them.

"Step (b)" Specify the known mass number of the measured standard sample from the CRT terminal and record it in the standard file. Note that since the time value and peak intensity at this time are unknown, 0 is recorded here. The target file is then given a name, and an unsupported standard file is created on the disk.

"Step (C)" Specify the standard file name and the number of the measurement data to be used from the CRT terminal.

"Step (d)" (a) Select 40 peaks from the highest intensity in the measurement data of the designated number.

(b) Furthermore, arrange the peaks in descending order of intensity among the 40 trees, calculate the theoretical mass number from the time value of each peak using several measurement condition elements, and store it together with the peak intensity.

The relational expression between mass number and time differs depending on the type of scanning, but in the case of the magnetic field type, M = k (H0 + αP) 2 ..... ■ In the case of the electric field type, it is Vo - αP. Here, M is the mass number, P is time, H is the offset of the magnetic field, α and k are positive coefficients, H5 is the value of the magnetic field at the start of scanning, and Vo is the acceleration voltage at the start of scanning. . In addition, the scan start mass number and scan end mass number of the measurement condition data are respectively M5 and Me, the time required for peak pattern scanning is T, and the sampling time is S.
Then, P is a time value in units of S.

Therefore, if there is a time value P of the current peak, and its theoretical mass number is M (then, from 2〇, in the magnetic field type, M = k (H, + αP) 2 M, = kHo2 M, = k (H0 + αT/s )2, so in the T/s electric field type, ■.-αP ■.-αT/s k H8′, so T/s ...... ■. Therefore, according to the type of scanning, the formula ■ If ■ is valid, the theoretical quality of the previous 40 peaks can be determined.

(c) Setting a tolerance of ±lO mass for the theoretical mass number,
Search to see if this mass number matches the known mass fi) number in the standard file. Note that the squares here refer to I a, m, and u. Compare the smaller standard mass with the 40 theoretical mass numbers that you have memorized first, and if i standard mass - theoretical mass number 1≦10.0, it will correspond to the standard mass. As a wax peak, assign the previous peak intensity rank to this standard square and write it down. Note that 10. in the above formula. O is a parameter and can be changed. When three standard squares to which at least one ranking has been assigned are retrieved, a relational expression between time and mass number is used to check whether the standard square has lost within a certain error range (variable with parameters). 3
If the numbers match within error, this becomes the estimated known mass number.

In other words, when there are three standard masses with m, < mz < m3, the time etc. of the peak expected to correspond to these (the peak whose theoretical mass number coincides with the standard mass within an error) is P, , PJ, Pu (P, <p, <P,)
Then, from the relational expression, in the magnetic field type, m, (P i + P o ) z (P J + P o )2, where, m: +/m+ 1, and in the electric field type, m3 - m.

is guided. The interpolated square m' is for the time value PJ, but how close m' is to the standard square m2 is checked within a certain error range (variable with parameters, usually 0.5%). The details are the following three conditions.

1m'-m, I O, 5 □ 〈 □ ・・・・・・ (11fly3ffL
+ 100 P, 4+Pa≧0 ・・・・・・ (2
)l M, -m, l < 5.0 ・-・
-(3) (Note that 5.0 is variable with parameters) Here, mo is the mass of the interpolation calculation result at time value O,
In other words, it almost coincides with the scanning start square. Moreover, the above (3) is a test to see how much difference there is from the actual scanning start square, and is a case where the absolute value of the difference is less than 5 squares.

m, P, ”-P, now three squares ml,, mz, ms (ml
Assume that the peak intensity rankings shown in the table below are assigned to <m2 x m3).

Here, ml is ranked 390 time value P3. When associated, m2 is associated with either P 3B% P 20 or P8, and m3 is associated with a time value other than the time value associated with ``I'', mz, and the conditions of the 3 formulas described above are established. If there is no combination that satisfies the condition for all ranks of m, find a standard square m4 that is larger than m3 and has rank assignment.Now m2, m3, A check is made on this combination using m4.

In this way, the algorithm searches for a combination that satisfies the conditions of the three equations in the order of m,..., mn+l, m11*2, so it happens that m,... is included in the standard sample measurement data. There is no peak corresponding to 1, and it is a different peak! If the standard mass-theoretical mass number 1≦10.0 is in the 11m range corresponding to m,..., naturally, m,..., m,..., 1, m,... The combination of 2 does not satisfy the condition, m7゜1, m,. 2. Moving on to the m7°3 combination. Also, m, , m□2, m
□You may check a combination of 3.

In this way, combinations of three peaks satisfying the conditions are found among the 40 peaks. Similarly, for known mass numbers other than the three previously selected known mass numbers, use the relational expression and appropriate error range (
(variable in the system) to create a standard file.

"Step (e)" Display the peak pattern of the measurement data of the specified number on the CRT terminal. In addition, near the top of the peak that corresponds to the known mass number, there is a special symbol (*) as shown in Figure 3.
etc.) so that you can visually see the response status.

As described above, the automatic calibration system for quality analysis data according to the present invention calibrates a predetermined number of high-intensity measurement data based on the measurement data of the standard sample, the measurement conditions, and the known mass number of the standard sample. The present invention automatically performs calibration by calculating a theoretical mass number for a peak and associating the measured data with the known mass number within an allowable error. Modifications are possible, and the present invention is not limited to the above embodiment.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, during calibration, the CPU automatically executes the work of specifying the known mass numbers and times for the peaks of the three trees, thereby reducing the time required for the calibration operation. be able to. In addition, in the new calibration, only the known mass numbers in the standard file are involved, and the mass numbers of 40 peaks with high intensity on the measured data side are estimated using the CPU, and the error is calculated by estimating the mass numbers of 40 peaks with high intensity on the measurement data side and determining the agreement between the estimated mass numbers and the known mass numbers. Since the comparison is performed within the same range, there are almost no erroneous correlations even in the measurement data of a standard sample that has many peaks with overflowing intensities.
It is possible to improve the matching accuracy.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the configuration of one embodiment of the autocalibration system for mass spectrometry data according to the present invention, FIG. 2 is a diagram for explaining the flow of processing by the autocalibration system, and FIG. A diagram showing an example of display output,
FIG. 4 is a diagram showing an example of the configuration of a mass spectrometry data system,
Figure 5 is a diagram for explaining the data stored on the disk, Figure 6 is a diagram for explaining the contents of the standard file,
FIG. 7 is a diagram for explaining measurement conditions, FIG. 8 is a diagram showing an example of the structure of measurement data, and FIG. 9 is a diagram showing an example of a peak pattern. 1...Measurement data, 2...Standard file, 3...
Peak detection section, 4... Mass number calculation section, 5... Assignment section, 6... Correspondence section. Applicant: JEOL Co., Ltd. Representative Patent Attorney Ryukichi Abe (2 others) Figure 9 (¥1 maki)

Claims (2)

[Claims] (1) An autocalibration system for mass spectrometry data that creates a correspondence table between peak appearance times in mass spectra and known mass numbers, and that extracts a predetermined number of peaks in order of intensity from measurement data of standard samples. Extraction means, mass number calculation means for calculating the theoretical mass number of the extracted peak, standard file in which known mass numbers of standard samples are registered, and peaks with theoretical mass numbers that are within the tolerance of the known mass numbers are assigned. It is equipped with a peak assigning means and a peak searching means for searching for a peak with a theoretical mass number that matches within a tolerance from the peaks assigned for three known mass numbers in order from the smallest mass number, and the appearance time of the searched peak. An automatic calibration system for mass spectrometry data, characterized in that the mass spectrometry data is matched with known mass numbers in a correspondence table. (2) The mass number calculation means calculates the theoretical mass number from the scan start mass number, scan end mass number, time required for peak pattern scanning, scan type, and peak appearance time during measurement. An autocalibration system for mass spectrometry data according to claim 1.